Regional Initiative: Which Appropriate Market Design?

Full text

(1)ROBERT SCHUMAN CENTRE FOR ADVANCED STUDIES. EUI Working Papers RSCAS 2009/60 ROBERT SCHUMAN CENTRE FOR ADVANCED STUDIES Florence School of Regulation. REGIONAL INITIATIVE: WHICH APPROPRIATE MARKET DESIGN?. Jan Moen.

(2)

(3) EUROPEAN UNIVERSITY INSTITUTE, FLORENCE ROBERT SCHUMAN CENTRE FOR ADVANCED STUDIES FLORENCE SCHOOL OF REGULATION. Regional Initiative: Which Appropriate Market Design? JAN MOEN. EUI Working Paper RSCAS 2009/60.

(4) This text may be downloaded only for personal research purposes. Additional reproduction for other purposes, whether in hard copies or electronically, requires the consent of the author(s), editor(s). If cited or quoted, reference should be made to the full name of the author(s), editor(s), the title, the working paper, or other series, the year and the publisher.. ISSN 1028-3625. © 2009 Jan Moen Printed in Italy, November 2009 European University Institute Badia Fiesolana I – 50014 San Domenico di Fiesole (FI) Italy www.eui.eu/RSCAS/Publications/ www.eui.eu cadmus.eui.eu.

(5) Robert Schuman Centre for Advanced Studies The Robert Schuman Centre for Advanced Studies (RSCAS), directed by Stefano Bartolini since September 2006, is home to a large post-doctoral programme. Created in 1992, it aims to develop inter-disciplinary and comparative research and to promote work on the major issues facing the process of integration and European society. The Centre hosts major research programmes and projects, and a range of working groups and ad hoc initiatives. The research agenda is organised around a set of core themes and is continuously evolving, reflecting the changing agenda of European integration and the expanding membership of the European Union. Details of this and the other research of the Centre can be found on: http://www.eui.eu/RSCAS/Research/ Research publications take the form of Working Papers, Policy Papers, Distinguished Lectures and books. Most of these are also available on the RSCAS website: http://www.eui.eu/RSCAS/Publications/ The EUI and the RSCAS are not responsible for the opinion expressed by the author(s). Florence School of Regulation The Florence School of Regulation (FSR) is a partnership between the RSCAS at the EUI and the Council of the European Energy Regulators (CEER), and it works closely with the European Commission. The Florence School of Regulation is sponsored by leading European energy companies. The objectives of the FSR are to promote informed discussion of key issues; to provide state-of-the-art training for practitioners; and to produce analytical studies in the field of regulation. It is a European forum dedicated to economic regulation. While its primary focus is on energy regulation, particularly in the electricity and gas markets, it is extending its coverage to other areas of regulation. This series of working papers aims at disseminating the work of scholars and practitioners on current regulatory issues. For further information Florence School of Regulation Robert Schuman Centre for Advanced Studies European University Institute Via Boccaccio, 151 I-50133 Firenze Tel.: +39 055 4685 751 Fax: +39055 4685755 E-mail: fsr@eui.eu http://www.eui.eu/RSCAS/ProfessionalDevelopment/FSR/.

(6)

(7) Abstract The European Union has a long experience and many success stories when it comes both to build a borderless Europe and to ensure that benefits are fairly distributed among producers and end-use customers. In some sectors results and benefits arise quickly, but sometimes borders remain difficult to cross despite numerous initiatives. A typical example of this is the completion of the single market for electricity. The process has been ongoing since the early 1990s and major progress has been made. However, we are still far from a borderless and truly competitive electricity market across Europe. A new legislative framework, the Third Package, will enter into force shortly and yield strong expectations. However, growing concerns become apparent among policy makers and in the market place on its ability to effectively foster the completion of the internal market and tackle market power issues. This paper argues that the approach adopted in the Third Package is not adapted to the challenges the European Union faces in electricity. The current lack of focus on implementing a better market design architecture leads the EU regulatory framework to overlooks important issues such as the promotion of power exchanges. The paper reviews the current state of the art on ‘smart’ market design in the economic literature and confronts it with the concrete experiences pursued at the regional level, in the European Union and beyond. Some of the issues discussed in depth include the TSOs’ roles and institutional design, generation adequacy and the design of capacity mechanisms and the development of demand-side response programs. It shows that the EU should learn from some of the on-going initiatives pursued at the domestic and regional level and that a sound market design based on a pool/TSO central dispatch is probably the way forward.. Keywords Market Design, Electricity, European Union, Regional Initiatives.

(8)

(9) Introduction The EU target of one single electricity market has been on the agenda since the mid-1990s but a real breakthrough is not only missing but still arguably far ahead. Both the 1996 and 2003 Directives had the best intentions with regards to removing trade barriers and ensuring efficient national electricity markets. Most national markets have been restructured and a better management of interconnection has contributed to increase electricity trade across Member States. More cross-country trade was also an outcome of the active participation of all stakeholders in new public organizations such as the Florence Forum1, CEER2 and ERGEG.3 These organizations have been a catalyst for this fruitful development, but the main target is still remote. The DG COMP Sector Enquiry (2007),4 together with the constant DG TREN’s monitoring efforts, identified the major shortcomings and barriers to the emergence of efficient electricity markets. The need to improve regulatory tools and remove market barriers was evident and urgent actions were needed. These findings initiated a major regulatory reform: the ‘3rd Package’. This initiative aims to fix all the ‘missing’ links in the current regulations and to create a pan-European regulatory agency called ACER. However, the proposal to impose full ownership unbundling of the Transmission System Operators (TSOs) encountered the opposition of several Member States.5 However, this major regulatory ‘update’ is still pending and, during the consultation process, the initial optimism has somewhat cooled down. Most elements of this new regulatory update are related to monopoly activities, but a much needed regulation specifically targeting more competitive markets is still missing. The latest update is on the new pan-European regulatory body, ACER. The recent developments indicate that ACER will enjoy very limited powers, which means that the objective of creating a strong European ‘FERC’6 has largely been abandoned. Creating coordinated power exchanges is still not an issue at EU level. Regional solutions for efficient congestion management, coordinated balancing markets and settlements emerge on a voluntary basis and are shaped by national regulatory traditions. The creation of an EU legislation to facilitate the regional initiatives is still not on the agenda. This paper tries to emphasize the main challenges remaining unaddressed in the 3rd Package. This contribution relies on the main findings of papers looking for a Market Design which will boost the development of efficient markets at the regional level. Most of these papers are based on concrete experiences of ‘good’ Market Design. The selection of contributions has been made exclusively by the author and is of course not neutral. It is the author’s hope that it can contribute to the debate on establishing better markets, both domestically and regionally.7 In the concluding part, a critical assessment is made in order to study how regulatory proposals are integrated into EU legislation. A critical issue is that a large number of prerequisites for wellfunctioning markets are not included in Directives or Regulations. One such example is the role and importance of power exchanges in ensuring efficient wholesale markets, as well as their role in the. 1. 2 3 4. 5 6 7. The European Commission initiated this forum as a ‘meeting place’ for national regulators, representatives of ministries and the main stakeholders in the power sector. Council of European Energy Regulators (CEER) Consultative body for the EC in the comitology process.. DG Competition, Report on Energy Sector Inquiry, SEC(2006) 1724 final of 10.01.2007. France and Germany were the main opponents. FERC is the US Federal Energy Regulatory Commission which- handles interstate regulation. The innovative discussion of Market Design issues took place during a relative short period of time in 2004–2006.. 1.

(10) Jan Moen. dispatch process. Voluntary agreements will not work when the parties cannot even agree on the opening hours of power exchanges! There are still a large number of issues addressed by EU law which are not efficiently enforced by national regulation and thus create some sort of regulatory ‘gaps’. DG COMP mapped several of these gaps and for the most part they served to protect incumbents from competition. The fact that these regulatory gaps still exist in spite of the major efforts of the EU since 1996 indicates the lack of coherence of energy policy and practice in Europe.. 1. The Current (Non-)Approach to Market Design in Europe Where did it all start – any cross-country lessons learned? The development of competition in the power sector took place in the late 1980s in an almost parallel way on two continents. Chile and Argentina in South America, and England & Wale and Norway in Europe, were the pioneers8 in the restructuring of the electric industry. However, the transfer of experience and knowledge between South American and European experiments was very limited.9 The main focus was on domestic issues. Power trade and regional initiatives were not on the agenda. The Argentinean reform mainly focused on domestic issues. The key elements of success were: spot market, a competitive structure in the generation market with no single dominant firm, nodal prices, well-organised dispatch, good investment incentives, etc. The absence of a fruitful dialogue with the other South American countries can be considered a limiting factor in the introduction of a broader debate on competitive power markets in Europe. In Europe, the regulatory system in England & Wales soon became the dominant model, although the early progress slowed down and dominant generation entities created obstacles to an efficient wholesale market. The rapid progress in Norway and the Nordic market attracted attention but a hydro-dominated system ‘far north’ was of limited interest to Central and Southern Europe. We note that the very few examples of regional initiatives took place mainly in hydro-dominated systems. Some of the first developments were in Norway in the late 1960s when the largest hydro generators formed an organisation to lower the risk of water shortage10 during dry/cold winters. Later on this organisation was transformed into a new pool named NordPool. When Sweden introduced competition in 1996, and by the same time joined NordPool, the first true international power pool was born. Very soon the other Nordic countries joined NordPool and the first organised regional initiative became a reality. The owners of NordPool were the Nordic TSOs. The long and fruitful Nordic cooperation in several other areas secured a smooth and long-lasting cooperation in the Scandinavian power sector. The net effect of power trade was evident and supported a trend of ever closer cooperation. This regional development was accomplished without any cross-Nordic legislation or regulation. This Nordic model was supported politically by the Nordic Council of Ministers, as well as by Nordel, the association of Nordic TSOs. The cooperation was complete with the creation of NordReg, the organisation of Nordic Regulators. A fully integrated governance structure11 was in place from very early on and contributed to a sound Nordic Market Design. It is important to 8. 9. 10 11. Chile and Argentina were the fist countries to introduce competition in South America and at the same time introduced some of the most innovative regulatory solutions. Stephen Littlechild published papers on the developments in South America, Argentina in particular, which have been innovative and of concrete relevance for Europe. Similar power exchanges were organised in the US and served the same goal. Each TSO had the same ownership part and equal representation in the board of Nordel.. 2.

(11) Regional Initiative: Which Appropriate Market Design?. emphasise the fact that the EU energy Directives did not represent a driving force; rather, an efficient Nordic market was perceived to be in the interest of all Nordic countries. In terms of governance and bureaucracy, this framework can be viewed as an independent, smaller, Nordic version of the EU. This Nordic model of Market Design was used as a benchmark when DG TREN wrote the draft Strategic Paper.12 This Market Design has passed several tests, of which the most severe was the situation in winter 2002–2003, when the hydro reservoirs almost ran out of water and spot prices ‘skyrocketed’!13 EU regulations and efforts to create regional initiatives The introduction of competition within EU Member States’ power markets had a very different rationale from that of the pioneering countries.14 Liberalisation and legislators thus faced different challenges and strategic elements. The Electricity Directives of 1996 and 2003 in fact did not introduce regulations and tools similar to the ones implemented in the Nordic and South American countries.15 Of course, elements from the Nordic Market Design could not be ‘blue copied’ into the EU legislation. But the Nordic model represented a fully coordinated Market Design covering the whole value chain from generation to retail sales, household customers included. The goal of restructuring was very much the same worldwide but the prescription, the ‘how to get there’, varied to a very large extent. The most basic element, the organisation of the wholesale market (NordPool), was organized first in the Nordic model, whereas the main EU building blocks were not introduced in the same logical order, and no clear path or direction was evident. Compared to the Nordic Model, the current EU legislation appears more fragmented and important ‘cornerstone’ organisations such as power pools are hardly mentioned at all. In order to facilitate trade, a new Regulation was adopted (1228/200316). This regulation covers the harmonisation of tariffs, congestion management and compensation for transit (ITC). The full transposition of this Regulation is still pending and important components have yet to be implemented.17 The idea behind this Regulation is to create new mechanisms to facilitate trade but it has not yet led to successful results. It is timely to quote the “lessons learned” from market reforms18 by Paul Joscow:. 12. 13 14 15. 16. 17. 18. •. Economic textbooks provide a sound guide for successful reforms,. •. Departure from economic textbooks is likely to cause problems,. •. Spot prices integrated with TSO operation are needed,. •. Spot markets need demand response,. The first draft had a clear view on power pools, but this element was not retained in later versions. See e.g. in the final version Strategy Paper – Medium Term vision for the Internal Electricity Market of 1.3.2004. Most contributors have made comments on this period. England & Wales (privatization), then Norway and later Sweden – and finally enhanced Nordic cooperation. At the fist Florence Forum meeting, Norway and the USA (FERC) were invited as main speaker. The USA had just showed interest in restructuring the power industry but the coordination of State Regulators with FERC was lacking and the final outcome was later the ‘collapse’ of the Californian experiment.. Regulation 1228/2003 of 26 June 2003 on conditions for access to the network for cross-border exchanges in electricity, O.J. 15.7.2003, L 176/1. Compensation for hosting transit (ITC), which still lacks a proper and operational definition, is often missing. Many stakeholders are still opposed to this arrangement, which may not promote trade and acts as a transaction-based mechanism. A full revamping had to be included as part of the 3rd Package but no efforts was ultimately made.. Joskow, “Lessons Learned from Electricity Market Liberalization”, The Energy Journal, Special Issue (2008). The Future of Electricity: Papers in Honour of David Newbery, 9-42. 3.

(12) Jan Moen •. Transmission investment is still a challenge,. •. Market power is a significant problem but the cure may be worse than the disease,. •. Strong political commitment to reform are essential,. •. Reform adjustment and fine-tuning will be needed, depending on the adequacy of the original design and its execution.. These lessons represent important issues when Market Design is set or revised. These bullet points highlight some shortcomings of the EU legislation, such as spot market integration and how adjustments or revisions of legislation should be conducted when changes are needed. The regional approach: What is missing? Do we have a proper Market Design? One of the most comprehensive studies of regional markets is the The Regional Approach in Establishing the Internal EU Electricity Market by Jacques de Jong.19 The report provides a good overview of the development both in the EU and in the US by focusing on the ongoing debate regarding the FERC’s initiatives on regional wholesale markets. The report emphasises that a large number of technical and political issues still need to be settled such as industry structure, market power, cross-border trade and integrating cross-border markets. The discussion will continue on many levels and the issues are numerous and complicated. The US is also subject to this ongoing discussion and challenges are very often the same such as regional market organisation and regulatory design. The report recommends the following: “to develop a legal framework for regional electricity markets with a set of minimum requirements. These requirements should include a physically and commercially strong interconnected system, a common view from the regulatory authorities on the prospects of regional markets, a clear and effective cooperative scheme between national TSOs and national regulators, a comparable level of TSO unbundling together with a set of mutually consistent rules for third party access (TPA) together with the intention of achieving integrated regional balancing markets and power exchanges […] And finally, the appropriate EU legal framework should be provided to establish the regional market procedures in a new EU regulation on regional electricity markets.”. To sum up the recommendations, it seems that a major revision according to a pool/TSO dispatch Market Design is the way forward. The current EU approach looks like a big ‘toolbox’. If progress is slow or insufficient, the box is opened and an ad hoc ‘repair’ is made. The repair very frequently represents a stricter and more detailed prescription for how regulation must be conducted case by case. This trend represents a sort of ‘continuous fixing’ policy which very rarely promotes proper incentives to create innovative and sustainable solutions. This paper presents instead a first step towards a smart EU Market Design, including suggestions for appropriate incentives. When this overall design has been set in motion, guidelines and more details can be added. Continuous small-scale repairs should not be the primary tool. The EU strategy has indeed perhaps reached a crossroad. Will stricter and more detailed regulations, rules and codes bring us closer to the objective, or will more support to incentives and a better Market Design be a better alternative? Markets do not happen over night – they need to be created and continuously supported in order to be competitive. Too many rules and regulations may hamper innovation and create barriers rather than. 19. Clingendal International Energy Programme (CIEP), December 2004.. 4.

(13) Regional Initiative: Which Appropriate Market Design?. opportunities to improve the functioning of markets. Fostering innovation and supporting new ways of thinking are often preferable to waiting for repeated regulatory ‘fixes’. Major updates needed – the revision of the 3rd Package CEER, ERGEG and EC all publish status reports, compliance documents, etc, to obtain feedback on how current regulations work and to discuss the need for updates and revised regulations. Combining these lessons learned and the critical outcome of the DG COMP sector enquiry, a major updating of the energy Directives was inevitable. This process of revision was labelled ‘the 3rd Package’, and it is in fact a major and comprehensive update of the 2003 Directive. The 3rd Package is targeting the missing links when it comes to crossborder trade monitoring and proposes to organise cooperation between regulators and EC (the new agency ACER in a more committed and stronger structure than the loose and bureaucratic ERGEG). TSOs will also be more strongly committed into the regulatory process through the establishment of ENTSO-E20 and its link to ACER. The 3rd Package represents a much more consistent approach than the previous updates, and power trade and regional initiatives are now high on the agenda. The creation of ACER as a separate agency for regulation will represent a substantial improvement, although the decision mandate is not very strong. A more distinct role for the TSOs and the new organisation ENTSO-E should contribute to mitigate the current shortcomings on calculating the optimum capacity of interconnection and to organise trade more efficiently. The 3rd Package has a strong focus on power trade. The revised legislation tries to remove barriers to trade and to provide a fresh approach to the coordination of Member State regulators, as well as obtaining a more efficient harmonisation of rules. Among the main inputs of the 3rd Package are the recommendations from the DG Competition power sector enquiry, which revealed a rather large number of barriers to competition. One important obstacle is the full ownership unbundling of TSOs, which is still a pending issue for the EC. As to the main arguments: without ownership unbundling and transparency, proper incentives for new investment will not be provided. The most important obstacle to make real progress when it comes to regional market development is the lack of a consistent Market Design across the EU. The current focus is still mainly on monopoly functions and too little attention has been paid to the creation of markets and how they can work efficiently. Several member states have a long experience of creating power exchanges which have contributed both to robust price settings and tools to better handling congestion management. This recipe is not integrated and coordinated well in the 3rd Package. The Strategic Paper proposed a role for power exchanges but pools are not included in Directives or Regulations.21 However, the obvious advantage of a power pool has become apparent and national or regional pools were created across Europe. This development was welcomed, but harmonisation and coordination among member states are still missing, and the full benefits have not been achieved. To compensate these imperfections, efforts were made to create second-best solutions, named ‘market coupling’, to improve the daily cross-border capacity utilisation among member states.22 20. 21. 22. This is a new and updated ETSO, which will coordinate and propose new codes on how TSO operations should be conducted. The strategic paper also proposed a timetable for the implementation. 2004: co-ordinated congestion management (combined TSO and PX coordination – implicit auctions); 2005: ITC into force and market-based solution in operation (Nordel Model?); 2006: access of all the Member States to a PX and congestion fully coordinated! Market coupling seminar in Brussels, ETSO, 9 April 2008.. 5.

(14) Jan Moen. These market coupling mechanisms have not always been successful23 and more robust solutions must be considered. The most efficient way to improve trade will be to coordinate and harmonise power pools to cover all regions of the EU. The need for harmonised and coordinated national/regional power exchanges becomes quite apparent when market participants try to benefit from regional markets and thus want to increase the volume of trade. Why the European Commission did bring these sound principles to a halt is still an unanswered question.. 2. A Full Debate on Market Design is Needed – ‘Repair Only’ Will not Work A sound Market Design for electricity was heavily debated in the USA in the early to mid-1990s. A Market Design based on the integration of a pool-based, short-term electricity market coordinated by a central dispatch operator provides a foundation for an open access system based on competition.24 This approach could be used to handle critical issues such as congestion problems and to ensure efficient technical operation, and at the same time be consistent with economic efficiency. To ensure optimal utilisation of the grid, the dynamic location of prices according to pool bids is necessary and contributes to the optimal flow of electricity. Some countries have successfully implemented basic elements of a design of this kind. The European debate on Market Design has so far mainly focused on security of supply and generation reserves. The baseline for most contributors has been that no market is perfect and that electricity is no exception. The main focus becomes to map market ‘anticipated’ failures and fix problems. An alternative strategy aiming to establish a robust design based on dynamic pricing, combined with central dispatch to optimise the use of resources, has usually been disregarded. A reopening of the Market Design issue in connection with the 3rd Package would have made sense but this opportunity was missed during the consultation process. In Europe, the Market Design concept has not been debated as frequently as in the US even though the SESSA project made some very useful contributions, such as the David Newbery’s Consensus on good Market Design (Stockholm, October 8, 2004). The David Newbery’s Market Design paper focuses on: •. Confidence in supply security. •. Sustainable competitive outcomes. •. Efficient free entry and investment. •. Efficient cross-border trade. •. Socially efficient emissions. David Newbery explains that the choice of Market Design is unlikely to offset poor market structures and avoid gaming while supporting integrated spot and interconnected markets. Newbery gives a warning on merger issues, especially when they take place between dominant gas and electricity companies. When Sweden introduced competition in the power sector in 1996, creating a common Norway-Sweden platform into NordPool was important. Vattenfall was indeed still overly dominant in generation and splitting the company in several entities could have been a possible solution. Sweden however had an international strategy of expansion for Vattenfall, who indeed subsequently acquired a strong market position for instance in eastern Germany through mergers and acquisitions. Enlarging market size through joining a coordinated power exchange was thus considered the best solution to mitigate potential abuses of market power.. 23 24. The latest example is Denmark – Germany. This argument was often used by William Hogan and papers published by the Harvard Energy Policy Group.. 6.

(15) Regional Initiative: Which Appropriate Market Design?. The Florence School of Regulation initiated a workshop on critical Market Design issues The Florence School of Regulation organised a workshop in July 2004: “The European Market for Electricity: Where Do We stand?” Some of the conclusions were: •. The capacity margin is lower than what is considered adequate.. •. Focus should be on price signals, investment and generation adequacy.. •. Experience shows that when markets are left to operate freely and send correct price signals, investors do respond.. The solution with respect to long-term generation capacity adequacy may be to introduce capacity support mechanisms. The debate on Market Design for this issue did split into two alternatives: •. As concerns whether competitive electricity markets provide sufficiently strong and early investment signals, the main focus inclines towards measures to repair market ‘failures’.. •. A pool-based, short-term electricity market coordinated by a central dispatch operator provides a sound Market Design. The main challenge is dynamic efficient pricing throughout the whole value chain in order to create incentives for efficient operation and investments.. Research about power Market Design in Europe has focused on the first point. The baseline for the first group was that the European Market has several features that veil or distort investment signals and incentives. Given examples are: no mandatory power pools, limited trade or trade only with neighbouring markets, demand side not sufficiently robust, demand management and real time meters not in place, limited cross-border capacity. Most European markets are capacity-constrained rather than energy-constrained. This sceptic approach has dominated the Market Design debate for the last 10 years, but no common large-scale capacity markets have been implemented. The advocates of the need for new capacity mechanisms often use the lack of demand response and inadequate generation incentives as arguments to focus on capacity markets. Various crises such as in California, Italy, New Zealand and Scandinavia are given as evidence to question the ability of competitive markets to provide proper investment signals. However, all these crises have been scrutinised and no common market failure has been identified. Most electricity legislations and regulations have addressed activities with natural monopoly elements, such as wires and system operators, to ensure transparency, low monopoly profit, quality and reliability. But the fact that the market cannot solve all problems on its own is an important insight that regulators must acknowledge and take into account. The main dilemma is how to combine the technical efficiency of the power system while implementing superior solutions from an economic point of view. The way forward will be to study the whole value chain carefully and decide which activities can be exposed to market forces and which ones must be left as monopolies in need of strict regulation. It will be important to obtain a full understanding of, and consensus on, why and how all these activities are linked. This Market Design approach may represent a major change to current EU legislation development and reshape energy policy for the future. Critical Market Design issues – some academic contributions The first paper is from The Faculty of Economics at Cambridge (2004)25 and was a joint project between E-Control26 (Austria) and NVE (Norway). The background of the initiative was the debate on ‘energy-only’ market and the introduction of reliability options (ROs) as well as their role in Market Design.. 25. 26. Newbery, Neuhoff and Roques, “Generation Adequacy and Investment Incentives in Liberalised Markets”, Faculty of Economics, University of Cambridge, 5 August 2005. E-Control is the national regulator in Austria.. 7.

(16) Jan Moen. The second academic paper, Electricity & Gas – Market Design and Policy Choices27 by L. de Vries, F. Correlje and P.A. Knops focuses on security of supply and Market Design. The paper underlines the weaknesses of energy-only markets and maps the current barriers to a better Market Design as well as generation adequacy shortcomings. By adding the Norwegian paper on TSO investment, a debate on generation adequacy and transmission investment will be possible. To illustrate a successful Market Design, Why the Nordic Market has Worked so Well? by Lars Bergmann is added at the end of the chapter. The introductory remarks for the Cambridge report underlines: “Opinions differ on whether liberalised ‘energy-only’ markets will deliver efficient, adequate and timely investment, or whether additional instruments, such as capacity payments, obligations or options are necessary and even desirable.” Poorly designed mechanisms could easily be counterproductive, distorting the market and leading to replacement of profit-motivated investment by tendered or system operator (SO)-contracted investment.28 Standard economic theory as applied to electricity markets (e.g. Caramanis, 198229) shows that a well-designed set of competitive (nodal) spot markets give prices that, if correctly predicted, would induce the efficient level and type of investment. The practical question is whether such market will function as predicted by the theory, whether investors will forecast future conditions and/or prices sufficiently accurately or whether they will be otherwise deterred (by risk aversion or the fear of regulatory intervention) from responding to those signals in a timely manner. There are subsidiary questions which are relevant in certain circumstances, such as whether generators with market power tend to over or under-invest (in base, mid-merit and/or peaking units) when the market is small relative to the size of generation units Similar questions can be raised in the specific context of hydrobased systems bordering a low-cost generation country and/or in the presence of significant (and subsidised) wind-power. As an oversimplification, whether or not competitive reforms included mechanisms specifically designed to ensure capacity adequacy depended on the initial conditions in each jurisdiction. In some cases (England and Wales, the Northeast US, among others) the vertically integrated utility had a welldefined planning margin for reserves, and it was considered important to protect this during the transition to competitive markets by a mechanism that would deliver the same degree of security. In other markets, continued state-ownership of the bulk of generation may have provided assurance that investment would be forthcoming when future margins would appear uncomfortably tight. In yet other jurisdictions, surplus capacity, concentration and the slow evolution of workable markets may have reduced the urgency to consider whether capacity payments were needed, and whether when created they might over-reward incumbents, given the fact that a satisfactory design would be challenging. Finally, there are important institutional differences between countries that may preclude some desirable solutions. If generation and transmission cannot be legally separated, or if there are many separately owned transmission systems, it may not be feasible to have a single TSO which manages dispatch and balancing. Instead an Independent System Operator (ISO) may be preferred. It is difficult to provide significant financial incentives to an ISO, while a TSO has adequate assets to bear the risk associated with such incentives, and makes the task of the efficient organisation of balancing, ancillary services and reserve procurement more straightforward. It follows that different countries may need to adopt different solutions, although it also seems reasonable that there should be some convergence to a 27. 28 29. A larger part of the paper seems to be based upon the dissertation of De Vries (2004): De Vries, “Securing the Public Interest in Electricity Generation Markets: the Myths of the Invisible Hand and the Copper Plate”, PhD Thesis (2004), Delft University of Technology. The theoretical part provides a good overview of the needs and design of capacity markets. Recent developments outline other solutions, however. Joskow, 2008. Caramanis, “Investment Decisions and Long-term Planning under Electricity Spot Pricing”, IEEE Transactions on Power apparatus and systems, Vol. PAS-101, No. 12, December 1982.. 8.

(17) Regional Initiative: Which Appropriate Market Design?. similar set of good practices, as countries become better interconnected and address similar issues of market power. There are thus a variety of questions that might be asked in response to how best to assure capacity adequacy. In a system that already has in place some mechanism to address a particular problem, such as a US style installed capacity requirement, or ICAP, the question may be how best to modify that mechanism. ICAP was the natural successor of the capacity obligations of the previous power pool rules after liberalisation. Only gradually was it perceived to offer a significant contribution towards financing new investment, and to offset the potential low average price resulting from price caps. In a market with extensive market power and an energy-only spot market with adequate current capacity, the question may be whether the market delivers generation adequacy, and how best to ensure that the market evolves towards a workably competitive or contestable state, while ensuring timely investment. In a competitive energy-only market (such as Britain) the question may be whether a mechanism is needed and if so, can be devised, to increase confidence in the delivery of efficient and timely investment without distorting the existing markets, or whether changes to existing institutions (such as the role of the TSO) can provide that assurance at a lower cost. The Cambridge paper is primarily directed to the situation found at present in the EU-15, which can be very roughly characterised as follows. Britain is a workably competitive energy-only market that has arguably still not been properly stress-tested as to whether new investment would be sunk on time if a future capacity shortage is predicted. Newbery discusses this case at length as it seems to present the most extreme case in which its market philosophy may be found inadequate. We argue that when the role of the TSO is appropriately defined and when an independent body is entrusted with producing high quality forecasts for demand and supply, it may not be necessary to devise any additional mechanism. The Nordic market is hydro-based, workably competitive, and has faced a serious drought in 2002/03.30 It is now contemplating new generation investment (in Finland in nuclear power, and in Norway in gas-fired capacity), as well as additional interconnection capacity (to the Netherlands). In their report, Henney31 and Bidwell (2005)32 cite a range of official views as to whether an energy-only market will be economically viable and whether it will be able to ensure generation adequacy: •. The British and Australians believe that it will, but they still have non-market back-up mechanisms.. •. The authorities involved in the Nordic market are debating (even though Denmark, Finland, and Sweden have back-up reserve arrangements), while the Dutch government hopes that adequacy will be assured by the system operator buying relatively short-term forward contracts to cover the peak of the forthcoming winter. Henney and Bidwell call their policies the ‘middle way’.. •. The Spanish, Irish, Argentineans, New Zealanders and the US FERC believe that a capacity payment is necessary to ensure generation adequacy.. It may be helpful to comment on these apparently differing approaches, to put them into context. First, the British and Australian approach is one in which the System Operators (SO) have a duty to balance the system in real time. In Britain, National Grid (NGC) as TSO has incentives to procure these 30. 31. 32. David Newbery does for the most part look at each single country in his comments but seldom includes Norway, Sweden, Denmark and Finland in a regional market approach. What ‘saved’ this market in 2002/03 was (i) close to optimum use of generation resources in the region and (ii) that high spot prices made a high import from ‘thermal’ EU member countries outside Scandinavia. Alex Henney made a project proposal on ROs to Norway, but NVE wanted a broader report on critical issues before such a ROs project would be considered. The Cambridge report served this purpose. See also Bidwell, “Reliability Options: a Market-Oriented Approach to Long-term Generation Adequacy”, 18(5) Electricity Journal (2005), 11-26.. 9.

(18) Jan Moen. balancing services at least cost and is allowed (indeed, even encouraged) to contract ahead if this reduces the cost.33 Australia has an ISO, NEMMCO, and a not-for-profit company managing the physical spot market and power system security over the entire national electricity market, NEM. To that extent, there is a less sharp distinction between these apparently energy-only markets and the ‘middle way’ described above. They are similar in relying on market signals feeding back from the balancing market to the spot, OTC and contract markets as the scarcity value of existing capacity is revealed in the various markets. In Australia, generators are paid a Value of Lost Load (VOLL), currently Aus$10,000/MWh, if the system cannot meet demand because of a shortage of generation capacity (but not, for example, because of a transmission failure). This should produce the same payments ex post than the capacity payments expected ex ante under the English Electricity Pool until its termination in 2001. In that market, generators were paid the VOLL times the Loss of Load Probability (LOLP) if declared available.34 The advantage for generators was that the capacity payments were made even if the LOLP was less than 100%, so they were paid more frequently, although at a lower value, thereby smoothing the revenue streams. The difficulty in the British context was that generators could game the system by not declaring all capacity available. The other markets all have special features that might lead to distorted signals for timely investment. The Spanish market was during the period of the Report overlaid with Competition Transition Contracts (CTCs) which were designed both to recover stranded costs and mitigate the substantial market power of Endesa and Iberdrola, who had about 80% of generation and supply, and faced little competition from imports (Crampes and Fabra, 2005).35 The capacity payments appear to provide a modest incentive both to be available and to bid into the voluntary Pool, OMEL, although it does not in any meaningful way reward scarcity. Argentina had a system of audited cost-based bidding into the wholesale spot market which clearly needed a supplementary capacity payment. Not surprisingly, as with the former English Electricity Pool, bidders could (and did) consider the combination of the energy and capacity bids when competing for space in the market, so the distinction between the two components was to some extent arbitrary. The Irish market is small and concentrated, generation is in a large part state-owned, and the market is in the course of an incremental regulatory reform with the North and South part of Ireland slowly integrating, possibly through interconnection with England and Wales. New Zealand is facing the problem of cheap, but unstable and only medium-term hydro storage and has yet to evolve toward a satisfactory solution. The US is a special but important case of a long history of regulated franchise monopoly utilities under private ownership, still governed by the 1935 Federal Power Act. The Act imposes a duty on regulators to intervene if necessary to ensure that electricity prices are “just and reasonable”.36 If a. 33. 34. 35. 36. National Grid Company merged with the gas transmission company Transco to form National Grid Transco, NGT, which is the holding company. We retain the original name NGC to stress its role in the British electricity system. This description of NGC’s responsibilities is perhaps an over-optimistic simplification of a still-evolving debate over NGC’s role in securing generation adequacy rather than system security.. The theory of VOLL-LOLP pricing implies that the ex post payment is the same as the expectation of the ex ante payment, but only if LOLP was correctly specified. Under the Pool the LOLP appears to have been systematically exaggerated (Newbery, “Competition, Contracts and Entry in the Electricity Spot Market”, 29 RAND Journal of Economics (1998), 726-749). Crampes and Fabra, “The Spanish Electricity Industry: Plus Ça Change…”, 26 The Energy Journal (2005), 127-154. This ‘reasonable’ regulation has introduced some market restrictions which limit the full transfer of experiences and knowledge to Europe. No restriction on prices at NordPool during the 2002/03 scarcity situation was an important component to overcome the severe hydro situation in Norway/Sweden. To avoid the use of market power – price caps are often used as well but this was not considered appropriate in Norway.. 10.

(19) Regional Initiative: Which Appropriate Market Design?. jurisdiction wishes to restructure its utility and de-regulate the wholesale market, it must provide FERC with evidence that the wholesale market is workably competitive, as competitive prices are by definition just and reasonable. If FERC is satisfied, it grants suppliers “market-based pricing authority” (Joskow, 2000).37 It is unclear whether markets deemed to pass this test would therefore be exempted from subsequent restraints (assuming no change in market structure through e.g. mergers), or whether a market which, for a given level of spare capacity, would deliver effectively competitive prices, but with lower reserve margin, would be susceptible to market manipulation and would force FERC to deem prices no longer “just and reasonable”. In addition, unhappy experiences with long-term contracting under the provisions of PURPA caused regulatory mistrust in anything other than spot markets as a measure of the wholesale price of power. This reluctance to contract ahead may have been compounded by the over-optimism of suppliers (retailers) who considered that spot prices would necessarily be lower than contract markets under competitive conditions. As Joskow and Kahn (2002)38 are able to demonstrate, given a reduction in supply from the Pacific Northwest, an increase in natural gas prices and a very sharp increase in the price of NOx permits in the Los Angeles basin in 2000, one would have expected the competitive level of prices to increase substantially in any case. Market tightness combined with a large fraction of demand being met in the spot rather than contract market provided the opportunity for generators to exercise market power and bid substantially above marginal costs (and the now-higher competitive benchmark), as well as withholding plant to force a higher market clearing price. Eventually, FERC was forced to intervene as wholesale prices were no longer “just and reasonable.” The Californian events that have so coloured reactions regarding the ability of liberalised markets to deal with scarcity can only be understood in that context. If regulators are now predicted to intervene when prices rise, then investors and banks are likely to be unwilling to invest in states with liberalised markets if they are allowed to price at variable cost during times of adequate capacity but not to earn the rents needed to cover capital costs in times of scarcity. The peculiarity of the US problem is that it remains a regulatory duty to ensure that market power is not unreasonably exercised while scarcity is adequately rewarded. Given the particular difficulty in an interconnected electricity system of distinguishing between cases in which prices are high because of genuine scarcity or because of market manipulation, it becomes attractive (and perhaps even necessary) to devise nonmarket mechanisms or obligations to reward scarcity. These may then be combined with varying aggressive market power mitigation procedures to deal with energy pricing.39 Jurisdictions (like the whole of the EU) which are not subject to such regulatory requirements start from a different position, and regulators and policy-makers should not be overly influenced by the special circumstances of the US. That does not necessarily imply that some additional mechanisms to reward capacity availability are unwarranted, but it does mean that the legal and regulatory environments are relevant for any such design. In addition to these institutional details, it may be important to distinguish between systems that are largely hydro (Norway, Austria, and to varying extents other members of NordPool), those that are relatively isolated or where imports are severely constrained (Britain, Ireland, Italy, Iberian peninsula). 37. 38. 39. Joskow, “Deregulation and Regulatory Reform in the U.S. Electric Power Sector”, in Peltzman and Winston (eds.), Deregulation of Network Industries: What’s Next? (Brookings Institution Press, 2000). Joskow and Kahn, “A Quantitative Analysis of Pricing Behavior in California’s Wholesale Electricity Market Summer 2000”, 23(4) The Energy Journal (2002), 1-35. Another concern of US investors is market power mitigation procedures involving bilateral contracts with must run generators and dispatch of generators out of the merit order without increasing the market-clearing price. They imply that generators in the same area might receive different prices for the same output. In the past, investors anticipated that longrun prices would be sufficient to finance new investment and would therefore in expectation also provide sufficient remuneration for today’s investment in future years. With differentiated pricing the reasoning might no longer hold – hence increasing the risk for today’s investors.. 11.

(20) Jan Moen. and those that are largely thermal or nuclear and well interconnected (most of the rest of the EU). The final dimension is the extent of market power, which is the ability of generators to manipulate the wholesale price of electricity from a competitive equilibrium price,40 but also includes the prevalence of vertical integration of generation and transmission. The extent to which vertical integration is a problem will depend on the efficacy of transmission regulation and the functioning of Third Party Access. Concentration in the supply of balancing and ancillary services can also distort prices and impede entry, reducing contestability. The Cambridge report underlines the role of the SO for system security and questions the role of the SO for system adequacy It is important to recognise that electricity requires a central SO to balance the system (or sub-system) in real time, and that any energy-only market must have such an SO function. The question then resolves into how that SO function is designed, and whether anything else is needed. Here it is important to distinguish between system security, security of supply and generation adequacy. System security requires the SO to balance the system in real time, if necessary by shedding load to prevent a rolling blackout. Security of supply is the ability of the system operator to meet short-run demand given existing capacity, if necessary by allowing imbalance prices to reach very high levels. Generation adequacy is a medium to long-term concept, and implies that there will be enough capacity available at each moment to guarantee security of supply at “reasonable” prices. Generation adequacy requires that investment in generation capacity is made in a timely manner to maintain an adequate reserve margin, and it is the main subject matter of this note. The required reserve margin will depend on the reliability of the existing generation stock and on the associated fuel supply, on peak demand uncertainty, on system-specific factors such as transmission bottlenecks, the amount of fast response reserve available, and on the pre-determined target reliability planning objective used by the SO. The SO in any electricity market must be charged to deliver system security, that is, to ensure demand and supply balance at all time. Ideally, this is achieved by ensuring adequate supply to meet demand at the relevant price (in the spot and balancing markets), rather than by shedding load. In many (possibly most) jurisdictions, the SO does not have any obligation to deliver generation adequacy, and thus may have to shed load if the imbalance price rises to its limit (a cap, if any, or the maximum offer that can be accepted). The requirement that system security and short-run balance is secured is thus not the same than ensuring either security of supply (no load shedding) or longer-term generation adequacy. This raises the central question of whether the natural approach to ensuring security and adequacy in energy-only markets is to require the SO to be responsible for these additional tasks. The critical issue with an SO approach to generation adequacy lies in the difference in time scales between managing the system in real time (security of supply) and contracting forward to induce sufficient investment to maintain an adequate plant margin. Such a step arguably represents a major change in the duties of the SO, and will need careful thought (and possibly secondary legislation or, at least, changes in grid and balancing codes and licence conditions).. 40. Market power can for example be measured by the proportion of time that the dominant generator is pivotal. A generator is pivotal when, without his capacity, the system cannot be balanced without shedding load. High import capacity not under the control of the dominant firm reduces the number of hours the dominant firm is pivotal. If two or more firms acting together are collectively pivotal, they may be able to tacitly collude to exercise market power. The extent to which they will be able to raise prices will depend on the elasticity of demand (likely to be low in most markets). Contracts reduce the incentive to exercise such market power, and ideally one would also wish to know the extent to which a generator’s un-concentrated supply was pivotal, but such information may not be readily available without a legal obligation on generators to reveal their contractual position to the regulator. 12.

(21) Regional Initiative: Which Appropriate Market Design?. Making the SO responsible for delivering generation adequacy efficiently thus requires three things: first, that the SO is legally responsible for maintaining generation adequacy (i.e. a predetermined reliability criteria); second, that the SO is properly incentivised to do so at least cost, and third that the SO is adequately credit-worthy to bear the risk of the contract position it may need to take. Britain provides a test case for the energy-only concept as the compulsory Pool with its capacity payments was deliberately abandoned and replaced with voluntary markets, a requirement to submit balanced schedules to the TSO, and a balancing mechanism to deliver system security, under the expectation that the model would also deliver security of supply and generation adequacy. This note therefore draws extensively on the British experience, recognising that other countries may also provide valuable evidence of market performance relevant to different circumstances. NGC would then have an incentive to find the most cost-effective way to maintain generation adequacy, be it to contract forward, or to propose any more specific capacity mechanism such as a ROs approach. The Cambridge report underlines also the importance of a balancing market that conveys scarcity signals In an energy-only market, balancing markets and markets for ancillary services such as reserve capacity have a critical importance for signalling scarcity. There exists a great variety of balancing mechanisms or markets in Europe.41 Principally, there are two types of imbalance price mechanisms: -. Dual imbalance pricing where a different price is applied to positive imbalance volumes and negative imbalance volumes (Britain, Poland, France, Sweden, Slovenia, Denmark, Netherlands and Italy).. -. Single imbalance pricing where a single imbalance price is used for all imbalance volumes (Norway, Germany, Luxembourg, Spain and Greece). A cost allocation method exists for Austria to allocate the cost of balancing energy to parties with an imbalance.. There are also two methods of determining imbalance prices: -. Average price of energy balancing actions (E&W, Poland, France, Denmark, Spain and Austria);. -. Marginal price of energy balancing actions (Sweden, Italy, Greece, Netherlands).. Market fundamentals dictate that during times of shortage, electricity prices should rise to the marginal cost of generation required to meet demand. One problem is that the marginal bid may substantially exceed the marginal or opportunity cost. Balancing mechanisms such as the British one which use an average pay-as-bid calculation for the price of imbalances combined with a dual imbalance price mute scarcity signals by paying generators their bid price and not the marginal price (in order to mitigate market power and possibly reduce volatility). But as Roques et al. (2005)42 noted, “[this type of design for balancing raises] critical issue in energy-only markets [that] lies in distortions introduced as prices feed in from the balancing market into the contract markets. In a multi-market framework such as NETA, the ability of investment signals to convey scarcity. 41. 42. The time resolution to which imbalances are settled is 15 minutes in Netherlands, Italy, Austria, Germany, Belgium, and Luxembourg; 30 minutes in England & Wales, and France; 60 minutes in Poland, Sweden, Norway, Denmark, Slovenia, Spain, and Greece. Roques, Newbery and Nuttal, “Investment Incentives and Electricity Market Design: the British Experience”, 4(2) Review of Network Economics (2005), 93-128. 13.

(22) Jan Moen. signals depends on the ability of the balancing price signals to feed in without distortion in the successive market layers characterised by different time scales. In Britain, the energy market is insulated to a degree from the costs of short-term balancing, both because of the average pricing formula and the lack of liquidity of the balancing mechanism. This insulation is most significant at times of scarcity, and creates the risk that the market will fail to deliver appropriate price signals for long-term investments. A price that might warn of impending shortage may indeed not materialise until the market is under severe stress, and the delay in the price signals might undermine timely investment decisions.”. The events of December 10 2002 on the NGC system are worth studying in this respect. On this day the system demand was the highest thus far recorded, and it exceeded the level forecast by National Grid. Whilst the price in the day-ahead market showed only a slight increase over the system peak to around £30/MWh, as well as the average System Buy Price (£71.6/MWh) in the Balancing Mechanism, the SO accepted offers in the Balancing Mechanism at £9,999/MWh for the marginal System Buy price. Price spikes and peaking units investment Efficient spot and balancing markets should give suitable price signals to indicate when and what kind of capacity is required in electricity markets. Such prices can be both volatile and extremely peaky, creating two related problems. The first is that generators may be unwilling to invest in risky peaking plant without contractual coverage. The second is that consumers may not hedge the risk of very high prices, and so may not provide the counter-party to a desired peaking contract or option, but may complain vigorously if exposed to high spot or imbalance prices. Experience demonstrates that peaking plants are likely to run a considerable number of hours per year. If the balancing mechanisms were changed to a single marginal priced market (as for example in the Netherlands) to reduce bidding risk and encourage bidding at marginal cost, and if peaking plants were then to bid competitively at avoidable cost, and if indeed as a result the peaking units ran for more hours, then it is likely that there would be both a greater convergence of spot and balancing prices and a lower average annual price of electricity. This brings us to the central problem of combining generation adequacy with competition. Factors affecting generation adequacy How far in advance construction of new plant needs to begin will depend on the type of plant, with combined cycle gas turbines (CCGTs) requiring the shortest time, typically less than two years from commissioning to operation (emergency diesel generation or open cycle gas turbine – OCGT - peaking plant may be commissioned even more rapidly.) Securing the necessary planning permits to build may take considerably longer, although building on existing generation sites (with grid connections, water for cooling and planning consent) may dramatically shorten this. How risky such investment will be will depend in part on the predictability of demand and future supply (which will be affected by the market structure and the extent to which other generators need to give notice of disconnections or new capacity plans). Uncertainty, risk aversion and ‘herd’ behaviour The impact of uncertainty about future demand on investment decisions depends on the considered type of plant. Uncertainty on demand can be thought of as either uncertainty over the average growth rate of demand (long term uncertainty) or as demand volatility in the short term. To consider only the extreme cases, the main criterion when considering an investment in a base load plant is the average demand growth, while investment in peaking units will be critically driven by the expectation of occurrence of high prices. While larger uncertainty over the long-term demand growth might. 14.

(23) Regional Initiative: Which Appropriate Market Design?. adversely impact investment in base-load units, greater demand volatility in the short run should make investment in peaking units more attractive.43 Most continental electricity markets remain fairly concentrated, so that the impact of uncertainty about future demand on investment decision has to take into account the impact of uncertainty on strategic behaviour. In a risk-neutral world, rational investors in a concentrated market confronted with demand uncertainty have to weigh two effects which work in opposite directions. The first effect relates to the value of waiting to get more information before committing to an irreversible investment. This encourages investors to delay investment decision. The second effect of demand uncertainty in an oligopolistic industry is referred to as the ‘pre-emption effect’. It captures the strategic advantage of being the first to invest. In the race to be first, investment may take place too soon. Demand uncertainty can therefore either speed up or delay the timing of investment for rational risk-neutral oligopolistic investors, depending on the relative strength of the two effects. Which effect is more important is therefore an empirical matter, as is the question whether there is a bias towards delayed investment that needs to be addressed by some mechanism to ensure timely investment. We return to this important issue below. This leaves open the question whether, given the best available forecasts, a liberalised market can be relied upon to deliver investment in a timely fashion in an energy-only market setting (supplemented by a TSO securing the necessary reserves to balance the system). In the past, with a franchise monopoly, there was if anything an incentive to over-invest for two reasons. The first reason is that when the investment is adequately rewarded (as under rate-of-return regulation or with abundant state funding) then more is better than less from the utility point of view (empire building or for the reasons set out in Averch-Johnson, 1962, and discussed below). The second reason is more defensible and comes from the perceived costs of being short (economic, but also political/regulatory) which are higher than the costs of being long (particularly if the capacity is paid for by captive consumers). In principle in liberalised markets the second benefit of adequate reserve is reflected in the higher revenue generators receive at times of scarcity. In practice, generators and society are risk averse. Neuhoff and de Vries (2004)44 argue that risk-averse investors put more weight on the bad outcomes – the years with low income – than on the sparkling profits in scarcity years and therefore reduce the equilibrium investment volume relative to risk neutral investors. Risk-averse governments and to some extent societies also put more weight on their bad outcomes – electricity shortages – and therefore prefer higher reserve margins than the liberalised market might deliver. This asymmetry created by risk aversion suggests that competitive liberalised markets might under-invest. Long-term contracts are the least interventionist approach to resolve this issue. They allow investors and consumers to hedge and eliminate the risk. Currently, policies required to support retail competition (such as the ability of consumers to switch supplier at short notice without penalty) undermine the ability of investors to sign such contracts.45. 43. Higher volatility of demand would normally translate into a higher volatility of prices. This means that the fraction of the year when prices are above average will increase, thereby increasing the profits of peaking plant which only runs in high price periods.. 44. Neuhoff and de Vries, “Insufficient Incentives for Investment in Electricity Generations”, 12(4) Utilities Policy (2004), 253-267.. 45. If consumers have to buy out longer-term contracts (as in the mortgage market), and if they express a preference for longer-term contracts, this problem might be avoidable. Evidence from mortgage markets suggests that consumers may have a preference for short-term contracts when prices are low, again introducing a possible misperception and market failure. There are also difficulties to align generators and consumers incentives in a long term contract. See on this, Finon and Perez, “Vertical Arrangements in Decentralized Electricity Markets: a Long Term Efficiency Perspective”, LARSEN Working Paper n°12 (2008). 15.

(24) Jan Moen. While risk-averse investors confronted with demand uncertainty might under-invest, suppliers contracting on behalf of their consumers are also likely to be risk averse given the high penalties they face in the imbalance market when they have under-contracted. However, under retail competition, the typical contracting time frame is measured in months rather than in years. Hence over-contracting by retail companies in the short time horizon for which they have to retain the tariffs fixed for final consumers may not give adequate and timely signals to assure the financing of new investment. Hydro systems and capacity adequacy There have been concerns that dominantly hydro systems like Norway may have particular problems to attract timely investment.46 There is some truth in this argument, essentially due to the complicated political economy of these systems. Typically, the dams and related transmission systems were built in the past and were financed by low (real) rates of interest. Their written down book values may now be very low, underwriting cheap electricity. This in itself deters new investment.47 Second, if the cheapest expansion option is more hydro capacity, then private investors may be reluctant to invest in a system that has high up-front financial costs but, given the near zero short-run financial cost, a tendency to set low average prices unless prices are set at the margin by conventional generation.48 To the extent that water management must take account of other non-electrical issues (river flow, irrigation demands, flood control, etc), private investors49 may be vulnerable to non-economic forces and pressure groups which are hard to predict and raise risk. Higher risk for capital-intensive dams is lethal as it raises costs one-for-one, as the average cost is entirely a capital cost. The threat of state or municipally financed competitive dams would further undermine private investment. In short, the investment climate for an inflow of private capital in hydro systems is generally unfriendly unless the average price of power is set by conventional generation. Fortunately, NordPool is increasingly well connected to neighbouring markets with nuclear and fossil generation. In long-run equilibrium, the average price in NordPool should therefore rise to equal the average price in neighbouring markets, which in turn will be the entry price for the most competitive generation needed, which with abundant storage hydro and adequate links would be baseload (although wind-power should be more attractive in hydro systems than fossil systems). The price characteristics of such a system are likely to be those of less daily price volatility, somewhat more seasonal volatility, and considerably greater annual volatility. That may or may not lead to a demand for multi-year contracting in order to smooth out the high price years (von der Fehr et al, 2005).50 In low price years the fossil generation will run less, and so on average over a number of years it is likely to have a lower load factor than for base-load generation in other thermal systems, and hence a higher average cost. In response, it may make sense for such generation to be supplied by lower capital cost and higher running cost plant (i.e. more like mid-merit plant). Equivalently, keeping older plant in the system for longer than might make sense in a purely thermal system may be the least-cost solution in a hydro system. On the other hand, the need of peaking capacity will be lesser so the annual demandweighted cost of electricity needs not be higher (and should be less, given that storage hydro allows for more efficient average utilisation). This is reflected in the current decisions to build base load. 46. 47 48 49. 50. In the hydro chapter, the Cambridge report may focus too much on country-specific terms and does not include the Nordel or NordPool co-operations which here represent a regional approach. In a normal year (as in 2009), Norway will operate under a surplus. In addition, a new big nuclear power station is under construction. A regional NordPool approach should ‘soften’ this statement. Recent investments do not fully support this statement however and this needs to be investigated further. In Norway private ownership in new investments is low. The government and the municipalities still represent a large majority. It remains to be seen whether new legislations will favour the development of private investment. Von der Fehr, Amundsen and Bergman, "The Nordic Market: Signs of Stress?", Cambridge Working Papers in Economics 0525, Faculty of Economics, University of Cambridge.. 16.

(25) Regional Initiative: Which Appropriate Market Design?. plants, like the 1400 MW of CCGT recently committed in Norway, or the construction of a nuclear power station in Finland. One would expect rationing (load-shedding) in some years, but at least in a hydro storage system this is likely to be more easily scheduled, particularly given the amount of energy intensive industry which may prefer to shut down. In short, it is not clear whether hydro systems face particular problems with resource adequacy, especially considering their different temporal dimension. This different temporal dimension makes it easier for regulators and policy makers to deal with it. On the other hand, hydro systems are exposed to larger year to year volatility of supply, and therefore energy shortage might be more severe than in thermal systems.51 Roques et al. (2005) conclude that “while electricity markets may be delivering adequate levels of investment, price spikes are testing government commitment to allow markets to sort things out.” In systems with significant reservoir storage capacity, there are two factors that should help give the government greater confidence to rely on market mechanisms to resolve capacity crisis. First, prices in a dry year are going to remain high during a relatively long period of time, but will be much less volatile than in a thermal system which is short of capacity. Thanks to arbitrage between reservoir capacity and energy, physical shortages are much less likely and emergency load shedding procedures easier to forecast and forewarn consumers. Second, policy makers under pressure to lower prices can point towards a rationale for the higher prices (the need to ration particularly energy-intensive demand) and a simple explanation (the lack of water in reservoirs). Such arguments appear to have worked in Norway in 200352 and limit public pressure from accusing deregulation or call for a temporary and inefficient Market Design fix. In addition, given the high proportion of electric heating in Norway, demand side responses were easier53 than in most European countries, with a higher fraction of higher value applications of electricity. Before highlighting and making some conclusions or recommendations on the Cambridge paper which is mostly focused on generation adequacy, the unbundling and investments paper of the Norwegian TSO investments paper should be scrutinized. Electricity & Gas – Market Design and policy choices54 At the same time than the Cambridge report, L. de Vries, F. Correlje and P.A. Knops made a report focusing on security of supply and Market Design. There are many overlaps between these reports on some of the same conclusions, but some important differences in the assumptions and reasoning are worth emphasising. The following statements are already well-known: power pools are not mandatory, some markets have significant traded volumes with neighbouring markets, market models (Market Design) vary greatly, and most of Europe is capacity-constrained rather than energy-constrained. But the main challenges are however: •. 51. 52. 53. 54. Why have power exchanges not enabled a larger dissemination and coordination when benefits are so obvious?. There is an important contrast with hydro systems where the least-cost expansion plan is more hydro, as in Brazil, and where the economics of thermal power needed for backup can be problematic. On the continent, with reasonable interconnection to markets whose prices are driven by thermal power, this is less likely to be the case. Professor Lars Bergmann supports this view and further comments are given in his paper “Why has the Nordic Market Worked so Well”. Demand response at the level of households and units in the service sector shows a 7% volume saving. But the aluminium industry was the main contributor, which far exceeded all other sectors. A larger part of the paper seems to be based upon the dissertation of De Vries (2004). The theoretical part provides a good overview of the needs and design of capacity markets. Recent developments outline other solutions however.. 17.

No results found